The existing research on the relation between undergraduate alcohol use and academic performance is fraught with design and measurement problems. As a consequence, there exists no strong policy rationale, based on academic considerations, for institutions of higher education to attempt to deter alcohol misuse among students. The proposed research project seeks to fill this knowledge gap to better inform public policy and to establish a strong rationale for addressing Healthy People 2010 national health objective 26-11b. This prospective study will examine the short-term relations between alcohol use and academic performance in freshman and sophomore students. The study will primarily rely on objective measures, rather than self-reports, to assess both late-night drinking and academic performance. Throughout each of the five semesters of the study, breath samples will be collected from distinct panels of students as they return to their residence halls between the hours of 10:00 p.m. and 3:00 a.m. on Wednesday through Saturday nights. In the first three weeks of each semester, students will be recruited for participation in the study. During this period, students will enroll themselves in the study using an access-restricted website. At the website, the study will be explained to the student, participation options will be identified, and informed consent will be obtained via a secure web page. The primary panel of students will be those electing to participate on a confidential basis. In exchange for financial compensation, the late-night intoxication patterns of these panels of students will be tracked over the course of one semester and linked to institutional measures of their academic work. The study will examine different ways to characterize nighttime drinking patterns based upon repeated BAL observations collected over the course of a semester. Students' academic performance will be weighted to account for course difficulty and course load. In addition, analyses will be conducted to determine whether college drinking mediates between academic performance in high school and that in college. That is, the study will test for possible compatibility effects whereby first- and second-year undergraduates select academic tracks that may accommodate their nighttime drinking practices. The proposed study will inform both public health and higher education policy by filling knowledge gaps about the nature of the relation between college drinking and academic performance.